{"id":64754,"date":"2026-07-30T11:38:06","date_gmt":"2026-07-30T14:38:06","guid":{"rendered":"https:\/\/manualdousuario.net\/?p=64754"},"modified":"2026-07-30T11:38:06","modified_gmt":"2026-07-30T14:38:06","slug":"human-made-badges-ai","status":"publish","type":"post","link":"https:\/\/manualdousuario.net\/en\/human-made-badges-ai\/","title":{"rendered":"I don’t really like the \u201cHuman-made\u201d badging trend"},"content":{"rendered":"

Two years ago, I wrote about one of those \u201chuman made\u201d badges<\/a> used to signal content free of AI. Since then, these badges have spread across corners of the internet hostile to generative AI.<\/p>\n

In a somewhat viral post from the indie web, Case Duckworth lists a few reasons to reject them<\/a>. His criticism isn’t aimed at the badges themselves, but at initiatives that centralize this kind of signaling. His view is condensed in this paragraph:<\/p>\n

<\/p>\n

However, I think that having a third party verify (or not, apparently) whether you\u2019re using AI in your personal site is endemic of the commodification of everything under late-stage capitalism. If you don\u2019t use AI on your site, just say so. Make a dumb button yourself, or use the one on my site (don\u2019t hotlink it, please!). It doesn\u2019t have to be this big production.<\/p><\/blockquote>\n

Later in the piece, he points out that some of the most popular badges (not by AI and Done by Humans) come from people who, on the side, are cashing in on AI-based companies.<\/p>\n

It strikes me as a parallel to Tools for Humanity, the company behind those orbs that scan people’s irises in exchange for cryptocurrency, owned by Sam Altman, who is also co-founder and CEO of OpenAI. These are questionable solutions to problems created by the very same people.<\/p>\n

I still hold the view that it doesn’t matter whether you use AI to write or not<\/a>, because the act shows through in the result. If it doesn’t, congratulations: the author made good use of AI \u2014 assuming that’s even possible; for those who oppose AI use on ethical or environmental grounds, the (lack of) quality in the text is beside the point.<\/p>\n

***<\/p>\n

On a related note, Substack is so overrun with AI-generated text that it launched a detection tool<\/a> to expose the use of the technology. According to co-founder and CEO Chris Best,<\/p>\n

The core problem is not people using AI, or the quality of its output. Not everything made with AI is slop, and not all slop is made with AI. The problem is when there is a mismatch between a reader\u2019s expectation and reality, especially when they unwittingly invest their attention in something with no human thought on the other end. That\u2019s Claudefishing.<\/p><\/blockquote>\n

(The neologism \u201cClaudefishing\u201d is a play on catfishing,<\/a> the term for pretending to be someone else in order to scam or deceive others. Nice one, Chris.)<\/p>\n

Substack’s new feature is a button that analyzes text and flags the likely percentage written by AI. It launched in partnership with Pangram, a leading company in the field. Which doesn’t say much, given that this kind of detection is inherently unreliable.<\/p>\n

That same Pangram published, in early July, findings<\/a> on the prevalence of AI-generated text across social platforms. Unsurprisingly, LinkedIn tops the ranking. Substack is where long-form AI-generated content is least common (10% of the total). If that’s already considered a problem worth this much attention and resources there, imagine everywhere else\u2026<\/p>\n

In rounding up reactions (well-curated to project a supposed balance, obviously), the comments showed no mercy<\/a>. It’s hard to find a single positive take on Substack’s new AI detector.<\/p>\n

Worth remembering: the same Substack going to all this trouble over AI-generated text is the platform whose CEO said, on camera, that it wouldn’t moderate explicitly racist content<\/a>.<\/p>\n","protected":false},"excerpt":{"rendered":"

In a somewhat viral post from the indie web, Case Duckworth lists some reasons to reject badges that indicate human-made content, without AI. In a related note, Substack launched a controversial \u201cAI detection\u201d tool.<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"link","meta":{"episode_type":"","audio_file":"","podmotor_file_id":"","podmotor_episode_id":"","cover_image":"","cover_image_id":"","duration":"","filesize":"","filesize_raw":"","date_recorded":"","explicit":"","block":"","itunes_episode_number":"","itunes_title":"","itunes_season_number":"","itunes_episode_type":"","_locale":"en_US","_original_post":"https:\/\/manualdousuario.net\/?p=64743"},"categories":[1575],"tags":[593,2055,2259],"_links":{"self":[{"href":"https:\/\/manualdousuario.net\/wp-json\/wp\/v2\/posts\/64754"}],"collection":[{"href":"https:\/\/manualdousuario.net\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/manualdousuario.net\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/manualdousuario.net\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/manualdousuario.net\/wp-json\/wp\/v2\/comments?post=64754"}],"version-history":[{"count":1,"href":"https:\/\/manualdousuario.net\/wp-json\/wp\/v2\/posts\/64754\/revisions"}],"predecessor-version":[{"id":64755,"href":"https:\/\/manualdousuario.net\/wp-json\/wp\/v2\/posts\/64754\/revisions\/64755"}],"wp:attachment":[{"href":"https:\/\/manualdousuario.net\/wp-json\/wp\/v2\/media?parent=64754"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/manualdousuario.net\/wp-json\/wp\/v2\/categories?post=64754"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/manualdousuario.net\/wp-json\/wp\/v2\/tags?post=64754"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}